Generative Artificial Intelligence Policy

Since the release of open source Generative Artificial Intelligence (GenAI) in late 2022, educational institutions have been challenged with a rapidly evolving form of technology that poses fundamental questions about knowledge, assessment and learning. A UNESCO study in June showed that fewer than 10% of the 450 schools and universities surveyed had developed policies or guidance on how to use Generative Artificial Intelligence.
During the Summer, Ecolint was present at the global launch of the 2023 Global Education Monitoring Report as well as participating in and following closely the UNESCO Digital Learning Week in September. This has allowed us to follow the evolution of global educational responses to GenAI and to have benefited from the help of experts in drawing up our position as a school. After several iterations and collaboration with specialists, our own technology partners and UNESCO, Ecolint has developed a policy statement.

Published
27 September 2023

Generative Artificial Intelligence Policy

Conférence des Directeurs Policy drawn up with the assistance of ICT campus partners and specialists from UNESCO - September, 2023

Preamble

The purpose of this policy is to describe Ecolint’s approach to Generative Artificial Intelligence (GenAI) within an educational scope. It can be applied to any age group where technology is integrated into learning. Given the rapidly evolving nature of GenAI, this policy will be reviewed frequently to assess any potential modifications that might be necessary in order to ensure that it remains aligned with our educational philosophy and values.

Fundamentally, Ecolint subscribes to the vision of UNESCO in its Guidance: GenAI should be used in educational settings to enhance learning, to improve our students’ understanding of the role of technology in the development and distribution of knowledge and to assess the limitations, pitfalls and dangers it brings with it. Technology is not where our thinking ends, it is where it starts.

GenAI allows students and teachers to access and synthesise information quickly, but it should not usurp critical and creative thinking and all educators must understand that since it works using statistical norms, that biases, overgeneralisations, simplifications and errors can be reproduced and magnified. Therefore, Ecolint teachers and students should always use GenAI with circumspection and criticality, remembering the more subtle and multiple facets of knowledge including minority voices.    

Above all, the use of technology in Ecolint learning spaces should always be screened through three fundamental questions teachers can ask themself when planning to create lessons and assignments:

  • How will technology be used to enhance human learning?
  • What are the ethical implications1 of this use of technology?
  • How will students reflect on the role of technology in their learning?     
     

GenAI, along with other forms of powerful technology can be used to improve teacher and administrator planning and assessment design. Following UNESCO’s guidance and this policy, this should be done in such a way that quality is improved and ethical boundaries (such as intellectual property, data protection, screening for inappropriate content) respected. 

Finally, GenAI needs to be situated within a larger matrix of competence development: Ecolint students are educated to be lifelong learners, to develop self-agency, to learn how to interact with others, with diverse tools and resources (such as technology) and with the environment in such a manner that they develop multiple literacies and transdisciplinary thinking. GenAI is a means to these ends and should be seen and used this way.

Using Generative Artificial Intelligence for Learning

  1. Gen AI, (Large Language Models [LLMs] such as ChatGPT, Bard, Alpaca, ChatPDF, Elicit, Perplexity, Chatsonic, Ernie and Claude - to mention a few) is

    an Artificial Intelligence (AI) technology that automatically generates content in response to prompts written in natural language conversational interfaces. Rather than simply curating existing webpages, by drawing on existing content, GenAI actually produces new content. The content can appear in formats that comprise all symbolic representations of human thinking: texts written in natural language, images (including photographs to digital paintings and cartoons), videos, music and software code. GenAI is trained using data collected from webpages, social media conversations and other online media. It generates its content by statistically analysing the distributions of words, pixels or other elements in the data that it has ingested and identifying and repeating common patterns (for example, which words typically follow which other words) (UNESCO, p.8)    
     
  2. It is understood that all learners2 have access to such technology, at home and, under the guidance and supervision of the teacher, at school.    
     
  3. Teachers are encouraged to use AI and GenAI in the classroom with their students (pending age restrictions)  to promote ethical, critical and creative use of this resource and to familiarise students with it so they can use it effectively in real life scenarios. Examples of this type of usage include using AI or GenAI to:
    1. find errors in computer coding or spreadsheet formulas;
    2. generate examination  responses and then asking students to assess the response against examination rubrics;
    3. find real world examples of abstract concepts/theories that can later be incorporated into student responses;
    4. build up a glossary of examples that can be used to illustrate claims;
    5. generate extended pieces of writing which can then be critiqued by individuals or groups of students;
    6. generate precise prompts that can be fed into GenAI and then to analyse the output critically and pedagogically.
    7. analyse, critically, images generated with GenAI and discussing how they could be improved    
       
  4. Teachers should discuss AI and GenAI with students to foster some critical appreciation of the manner in which information is gathered and distributed by it so as to understand the nature of algorithmic meaning-making, including the mistakes that it can make3. These conversations should be curated according to the age of the students.    
     
  5. Students can reference AI and GenAI as a source in their written and oral production but must ensure, as a general principle, that they refer to a variety of sources and not just AI or GenAI.4    
     
  6. Every year, ICT campus partners will run a series of training sessions for teachers on how to best integrate GenAI into learning. Much of the substance of these trainings will draw from the UNESCO Guidance for Generative Artificial Intelligence (a resource which is expected to evolve).    
     
  7. When teachers allow students access to the web in their learning, it is understood that this includes GenAI, pending age appropriateness.    
     
  8. The spirit of using AI and/or GenAI is for it to stimulate creative, ethical and critical thinking, but thinking that is the student’s and not to simply generate an entire answer or product using AI or GenAI so as to substitute student thinking with machine responses.    
     
  9. Summative and external examination based assessments completed in non-examination based conditions must be routinely screened5 for their AI generated content.  Students should be made aware of this by their teachers, heads of years or academic assistant principals.    
     
  10. School academic honesty policies should be regularly reviewed to reflect the opportunities and challenges of AI.6    
     
  11. Any action that requires students to create accounts with digital services, including Generative AI platforms, must be done with a discussion with the ICT Campus Partner.

 

Sources

Giannini, S. (2023). Generative AI and the future of education. UNESCO. 
Miao, F., Holmes, W., Huang, R. & Zhang, H. (2021).  AI and education: guidance for policy-makers. UNESCO. 
UNESCO (2023).  UNESCO Guidance for Generative Artificial Intelligence    
 


1 For example: intellectual property, data protection, screening for inappropriate content    
2  Note some GenAI ( chatgpt, Bard, Claude) have an age requirement to access their interface: usually 13 years old although this might vary by country.    
3 Ethics and issues of gender and racial bias would also be an important part of these discussions: https://www.ajl.org.    
4 Some reference systems require citation of AI sources (see https://style.mla.org/citing-generative-ai/,  https://apastyle.apa.org/blog/how-to-cite-chatgpt and https://camosun.libguides.com/ai/citation).    
5 There are a number of detectors (https://www.turnitin.com/solutions/ai-writing) but in tandem there are numerous free easy ways to circumvent these such as https://undetectable.ai/?via=t9v9g. This is an important creative tension for staff to unpack and discuss with students.    
6 This is the responsibility of the Director of Education.